Gastrointestinal stromal tumors (GISTs) are associated with a dismal prognosis in localized and advanced phase with a major resistance to conventional chemotherapy agents. Virtually all malignant GISTs actually harbor activating mutations of the KIT pathway in the tumor cells, leading to ligand-independent activation of KIT tyrosine kinase activity and tumor growth in vitro. Glivec® inhibits KIT and exerts a major antitumor efficacy in vivo in patients with advanced GIST. Glivec® is generally pursued until progression or intolerance. The optimal duration of treatment with Glivec® remains unknown. The objective of this study is to determine the feasibility of Glivec® treatment interruption with reintroduction at progression in GIST patients.

Gastrointestinal stromal tumors (GISTs) are associated with a dismal prognosis in localized and advanced phase with a major resistance to conventional chemotherapy agents. GIST cells are positive for KIT (CD117) and CD34 in 100% and 70% of cases, respectively. Virtually all malignant GISTs actually harbor activating mutations of the kit pathway in the tumor cells, leading to ligand-independent activation of KIT tyrosine kinase activity and tumor growth in vitro. Glivec® inhibits KIT activity at an IC50 of approximately 100 nM which is similar to that required for inhibiting the tyrosine kinase associated with Bcr-abl and the PDGF receptor. Experiments on cell lines containing an activating juxtamembrane mutation (similar to that found in GISTs) and cell lines containing transfected wild type KIT protein, showed that these cells appear to be strongly dependent upon the activity of the mutant receptor to prevent apoptosis, thus providing further scientific justification for the development of Glivec® as an antineoplastic agent with specific activity against GIST as a KIT-driven malignancy.

Since the first single patient with metastatic GIST treated by Glivec® in March 2000 (16), more than 2000 patients have been included in prospective trials testing activity and tolerance of Glivec® in patients with advanced/metastatic GIST. High response rates have been documented, only a limited percentage of patients progressed after achieving objective response, and median survival has not been reached in all studies. There has been no clear demonstration of a dose-response relationship. About 15% of patients experienced a rapid disease progression under treatment but the mechanisms of resistance remain unknown. Some patients progressing at 400 mg/day further responded to higher doses of Glivec®. Toxicities were infrequent, mainly mild to moderate and their incidence seems to be related to the total daily dose administered.

The optimal duration of treatment with Glivec® remains unknown. In addition the impact of surgical procedures of tumoral residual masses is not yet evaluated on progression free and overall survival. The objective of this study is to determine the feasibility of Glivec® treatment interruption with reintroduction at progression in GIST patients.

Primary objective

- To compare progression-free overall survival beyond 1 year in patients treated by Glivec® achieving a CR, PR or SD at 5 years. Patients will be randomized between 1) interruption of Glivec® until progression w/ RECIST criteria and then re-start (group 1) vs 2) maintenance of Glivec® (group 2).

Secondary objectives

• To compare overall survival in the two groups of randomized patients.
• To determine progression free survival beyond 1 year in patients in CR, PR or SD at 5 years who refused randomization and 1) selected Glivec® interruption or 2) chose Glivec® maintenance.
• To determine CR, PR and SD rates after re-start of Glivec® in group 1.
• To assess the number of patients who completed radical surgery on tumour residual masses after an objective response.
• To assess resource utilisation by evaluating direct and indirect cost.
• To evaluate the correlation between the serum rates of Glivec® and the response to the treatment w/ RECIST criteria, in patients with the diagnosis of GIST and treated by Glivec® 400mg /day, and this until progression, stop treatment or study exit.
• To follow immune modifications induced by Glivec® administration potentially related to clinical response and toxicity.
• To realize if possible the sequencing with aiming diagnoses KIT, in order to evaluate the correlations existing between the response/or the absence of response to Glivec® and the type of mutation of KIT.

Overall study design : This is an open label clinical trial of oral Glivec® 400 mg/day in a population of patients with metastatic and/or unresectable malignant GIST in relapse. 564 patients will be enrolled in ten years in 20-30 French Cooperative Centers.

Treatment : Patients will receive Glivec® 400 mg /day for an exposure period of 60 months. At the end of a 5 years period, patients with non progressive disease will be proposed for randomization between 1) interruption of Glivec® until progression w/ RECIST criteria and then re-start vs 2) maintenance of Glivec®. Patient who refuse randomization will be proposed either solution and followed according to the same schedule. During treatment with Glivec® 400mg/day, Glivec® may be increased to 600 mg/day or 800 mg/day if the patient is progressing. In case of re-progression, the patient will be excluded of this study.

Signed informed consent for the study, including the possible randomization, will be obtained.

Inclusion Criteria:

1. Patients 18 years of age or over.
2. Histologically documented diagnosis of malignant GIST.
3. Immunohistochemical documentation of c-kit (CD117) expression either by the primary tumor or metastases using the DAKO assay.
4. Performance status 0,1, 2, 3 (ECOG)
5. Adequate end organ function, defined as the following: total bilirubin < 1.5 x ULN, SGOT and SGPT < 2.5 x UNL (or < 5 x ULN if hepatic metastases are present), creatinine < 1.5 x ULN, ANC > 1.0 x 109/L, platelets > 100 x 109/L.
6. Female patients of child-bearing potential must have negative pregnancy test within 7 days before initiation of study drug dosing. Post menopausal women must be amenorrheic for at least 12 months to be considered of non-childbearing potential. Female patients of reproductive potential must agree to employ an effective barrier method of birth control throughout the study and for up to 2 weeks (according to updated Invest. Brochure) following discontinuation of study drug.
7. Written, voluntary, informed consent.

Exclusion Criteria:

1. Patient has another malignant tumor in CR<3 years (except if the other primary malignancy is inactive and not requiring active intervention). Previous basal cell skin cancer or a cervical carcinoma in situ are allowed.
2. Patient with Grade III/IV cardiac problems as defined by the New York Heart Association Criteria. (i.e., congestive heart failure, myocardial infarction within 6 months of study)
3. Female patients who are pregnant or breast-feeding.
4. Patient has a known diagnosis of human immunodeficiency virus (HIV) infection.
5. Patients received chemotherapy within 2 weeks prior to study entry, unless the disease is rapidly progressing
6. Patients had a major surgery within 2 weeks prior to entry study
7. Patient with any significant history of non-compliance to medical regimens or with inability to grant reliable informed consent.
8. Previous treatment with Glivec®